I've been looking into the heatsink question for this board recently, and
was hoping to get some parts to check with before commenting; but since this has appeared now :shrug:

The heatsink mounting hole pattern appears to be within 0.025" (0.635mm) of that used on NV40 reference graphics cards. i.e. each hole is 0.318mm away from ideal, or in other words about 1% innacurate. This should probable be taken care of through tolerances (this is one of the things I wanted to check).

The bottom line is that I think any graphics card cooler for the Nvidia 6800 or above (including the Vf700) should have a mounting mechanism in the correct place. However, this doesn't mean that there won't be obstructions (another thing I wanted to check).

Since I won't get any of the parts I'm waiting on for several weeks and since I broke my wrist couldn't do the tests I wanted to at the moment anyway here is something that hopefully is helpful.

I received my i915Ga-HFS today and it has the same hole pattern as the i915GMm-HFS. I guess there's more space for a larger, modified heatsink around the processor, though, and the pci-express slot is far, far away from the processor. My friend recently bought a vf700 for his video card (I think he has it mounted already), but I'll see if it looks like it will fit on the processor, as Romeo suggested.

As promised, I finally got my Borg HPC 479 heatsink compatible with this Aopen board from www.mcubed-tech.com.

First impressions is that the thing is heavy!! After installing it very carefully over the CPU, I proceeded to the mounting of the heat pipes. Not easy to find a space in the case to attach the other end of them ( see instructions at their site) so after much bending and twisting, I was able to install 3 of them. Since it is supposed to ran fanless, I tried just that with my case open. Note that I tried just some basic tests like running a DIVx movie at +- 15% CPU usage. This gave a stabilized temp of 54°C.

After that, I decided to rip the Aopen fan from the original Aopen HS and put it on top of the Borh HS (just deposing it, no fixations). Same test gave a fluctuating temp between 47°-52° running the fan at 50% RPM (lower than that the PC would starting beeping. Could it be that the fan simply stops spinning?). Since at this level the PC is nearly silent, I decided to stick with this combination.

Sorry about the summary tests but I don't have that much time at my disposition to do the full thing. My final conclusion is that the Borg is sufficient to calm my nerves about this Aopen MB temp issue, albeit at a price.

The main use of the machine is MCE 2005 and to record TV. Most of the time is in standby mode.

Hi dragonii,

Haven't followed this thread lately, but I thought I'd let you know that I too
have the mcubed case, the BORG Pentium M heatsink and Aopen's ATX i915 p-m board

I haven't yet installed mine, but I do notice that there are a lot of "manual" steps involved with the pipes, the contact quality with whatever thermal paste you use, etc. I kind of expect better temps than what you are reporting, so when I set mine up I'll let you know. That way we'll be able to see how much tweaking may be required for optimal performance.

Your temps are fine though for the P-M, so I'd be happy with that.

http://www.compuvest.com/Description.js ... iid=138248


Check out that link.... looks interesting (only the Japanese product description is available on the cooler master site, but it says it's about 50mmx50mm, so that is probably the right size.

I have my 2.0GHz P-M on the ATX version of this board, and the temps are quite strange. It usually ends up idling at around 50C, maybe a little lower, but after playing a very graphics-intensive game (i.e. Battlefield 2) or something like that, it doesn't go above 65C, and usually stays at around 60.

I might end up getting that CM heatsink, but I definitely do not trust the 25db rating.


EDIT: Better yet, how about something bigger? I found anothing interesting heatsink: http://www.sundialmicro.com/Silverstone ... 7_491.html

It says it supports socket 479... even the picture seems to have a bracket that can be used for P-M...

I got my mCubed case and AOpen i915GMm-HFS board last week, and installed it last weekend.

First problem was that the SilverStone ST30NF power (or more specifically the mains power cable) that came with the case interfered with the SATA3 and SATA4 connectors of the board. There was no way to install both at the same time, so I ended up tilting the PSU somewhat so I was able to fit them both. Looks like my "custom" PSU installation works fine, though.

I only installed 2 of the 4 heatpipes for the CPU (as it looked to me that the CPU end of the heatpipes was higher up than the end that goes to the side of the case for the two lower pipes, which makes the heatpipes inactive if I correctly understand how they work, and as I only use Pentium M 760 (2.0GHz) processor).

I also mounted one of the other two heatpipes on the north bridge (and to the case side below where the CPU heatpipes are connected), as the north bridge seemed to become much hotter than the CPU. After the heatpipe installation the north bridge felt cool to touch (instead of almost burning).

My idle temps are around 41 degrees C, and after 10 hours of running at 100% CPU encoding an XviD, the CPU temp is at around 54 degrees C. I am pretty happy with that, I was actually expecting higher temps with only two heat pipes connected.

Pate

Ralf, I don't know if you're still following this thread, but do you remember what the 2-D quality was of the onboard DVI. Would you say it was comparable to other digital output?

Yes, it was as good as the DVI out that I'm used to seeing with my various ATi cards.

Actually, I've seen virtually no difference in the quality of any DVI source. They all seem very similar and are all miles ahead of good 'ol analogue.

Thanks, Ralf.

A quick question: Has anyone been able to successfully run all the way through 3DMark05 using this mb in a system with the PCIe x16 slot filled? Does anyone use it regularly for games? It's less of a silly question than you think: I've yet to get by the first scene in this test with this board, a 760 (2.0GHz) and a 6600GT in the PCIe slot.

It'd be nice to know that someone can use this board for gaming. Thanks.


-j

Ralf, in the main review, you describe the fan as 40mm, but in the postscript it says 50mm. Please clarify. Thanks.

Man am I glad I found this place....

I realize the topic is currently about temps & video, but this is the only place I know to ask this....

I recently built one of these up and I can't get below 76 watts at idle (at 800Mhz, .98v). I don't get it. It pulls 15 watts, while it's OFF!

P-M 760, i915GMm-HFS, 2x 512 DDR2, momentus 5400.2 sata, plextor px-716SA, zalman 7000 w/ nexus fan that barely ever kicks on, and a phantom 350 (that is supposedly efficient?)

Did I miss a bios setting or something??

That's very wierd.

What are you using to measure power usage? Are you sure it's reading correctly?

If you have something with a known power draw (a lamp, for example, with an xx watt bulb), try measuring that and see if the meter is giving you the correct reading (should be close to the wattage of the bulb).

Yeah, its definitely really strange.... I'm using a decent (Extech) multimeter, tested with a lamp that has a 60 watt bulb. The lamp came out to around 57 watts.

The multimeter is hooked up, intercepting the hot wire (the one running from the upper right prong of a standard socket), and readings were taken during bios (95w), windows boot (82w), idle windows desktop (75w at 800mhz .98v & 88.8w at 2.0Ghz 1.356v), and sandra burn in w/ cpu arithmetic (102w at the 2Ghz setting).

Also, double checked the results with a Fluke 337 clamp meter.

Wall voltage was 118.5. An example, the desktop idle at 800mhz, .98v, came out to be .64A W=A*V right? .64 * 118.5 = 75.84w.

Is that the proper way to check? Thanks for the help, I still can't believe it came out this high. I'm going to start unhooking things soon and seeing if that helps.

Hmm, everything looks right. You're using the onboard VGA, right? I'd expect numbers like those if you were using a high-end video card, but not with integrated video.

Yeah I'd say start unhooking stuff and see where that gets you. Let us know what you figure out, I'm really curious to see what's causing the problem.

Since you have a clamp meter, you could also try measuring current across the power supply wires on the DC side, maybe it'd be helpful to know how much total power is being drawn on the DC side (so you can tell if maybe you got a bad Phantom with poor efficiency) or how much power is being drawn on each rail?

Yeah, I'm using the onboard VGA via the DVI hookup.

I tried unhooking the hard drive and dvd already, and booting into the bios. The current draw on the AC side is the same as with them hooked up, (around 90 watts). I tried disconnecting the mobo's connectors (the 20pin, and the 4pin) and switched the power supply on. The 15 watts it had been pulling while off went away, down to zero. So that was definitely coming from the mobo. Next I took out the ram and switched the power supply back on (not turning anything on, only applying power to the motherboard in its 'off' state with the small red light that comes on.) Still pulls its 15watts.

So its not the HD, DVD, RAM, and the CPU fan is off most of the time. So it has to be the mobo, cpu, or power supply. I'm thinking its the power supply. While buildign this thing, i got carried away and did the sleeving on the power wires... So its going to be fun trying to get that clamp in there!

Btw, thank you for all of your help.

Your PSU does not have Active PFC. This means the current draw from the AC outlet is much higher than it would be with APFC. You are measuring VA or apparent power with your setup, not W, which in AC are not the same. VA does not reflect the true power (watts) that the PSU is actually drawing.

Go to the Phantom 500 review. In the results table, you will see that power factor is 0.57 at 65W output and 0.58 at 90W. This means that the PSU draws apparent power of 114VA in order to obtain true power of 65W.

If we assume that Power Factor is 0.57, then your 75VA indicates true power draw of 43W. This correlates pretty closely with RH's Pentium-M system power draw measurements -- and my own -- using a true power meter such as the Seasonic Power Angel.

WOW

Well, I was going to bed.... but it looks like I have a lot to learn about electricity!

Thanks for the knowledge-bomb, that definitely helps out! I've got a LOT of reading to do.

Yeah, thanks for the info Mike. I had no idea.

IIRC, the included fan is 50mm so the original 40mm comment falls into the "typo" catagory.

You're welcome.

There a brief explanation of Power Factor in the PS Fundamentals & Recommendations article -- page 5. Lots of info on the web.

I've read alot on PSU's, AC to DC efficiency, and power factor, but one last note. I understand that this method will tell me the actual power that the computer's components are consuming from a system standpoint, but as far as my electricity bill goes, I am still consuming 75watts from the wall though right?

I'm still reading up on all this. Sorry if its a dumb question I'm getting hung up at the difference between efficiency and power factor.

In other words, even though my systems components are using 43W, my house is basically seeing the equivalent of something like a 75W lamp being plugged in?

It depends on how your power company bills you. As a general rule, private homes in North America get billed for kWh. With this kind of billing, they bill you for watts, so you'd pay for the 43W. Industrial customers usually get billed for VA, so they'd pay for the full 75W.

same deal for AU. (fill out your location in profile )

i've always wondered why there is this incentive for large organisations to clean up their PF, but not for residential. it seems like it would be in the power company's interest to make everybody have corrected power factor, suburbia inclusive..

Probably because residential users simply don't have much choice or control over PF. They buy electrical goods that are sold. If they happened to have poor PF, how would they change it or even know?

Just wanted to say thanks to everyone. See what gets started when I focus my attention outside of a computer tower for 5 seconds!

I don't deal with AC electric much as you can tell. Found a good write up on PF on page 3 of this APC document, for anyone that may be interested.

ftp://www.apcmedia.com/salestools/SADE-5TNQYL_R0_EN.pdf[/i]

Thanks for this nice thread and review (that maked me decide to buy it..)

Just got my Aopen i915GM-HFS rev2 yesterday and plugged in my 2.13Ghz Pentium M processor.

To my surprise the CPU overheats to 112 degrees in a matter of seconds..!!! the system shutsdown automatically..! probberly to prevent hardware loss..

What did I do wrong here..? I mounted the heatsink with a drop of compound pasta.. hold it steady with one finger while cross tightning the springs.. the fan is spinning at 100% (>3500rpm)

The motherboard is mounted in a Thermaltake case with 1 x 80mm fan and 2 x 60mm fans. The PSU is a NorthQ 400W with a 120mm fan.

Please help me out here..

Thanks ;o)

____________________
(\/) Y S T A L

Well, let me be the first to point out that pasta doesn't conduct heat well

Are you reading the temps from the bios? And, what heatsink do you have, the small one that originally shipped or the larger (yet too small, if you ask me) one that shipped later? And what thermal interface are you actually using?


-j

The "pasta" heatsink compound I use is included wih the Aopen heatsink. The heatsink is the new "bigger" one

Anyway, i tried to put my thumbs on the heatsink while reading the temps in the bios.. and it suddently dropped down to 65 degrees. As soon as I let go, the temp gets back at 100+ again.

I then tried to lower the voltage on the CPU to 1.1v and this keeps the temperature down to 77-83 degrees (and below 56 when putting pressure on it)

I also tried to reasamble the heatsink for the 10th time, just to be sure that the bolts are tightned well and evenly.. but that didnt help either

Any suggestions before sending it back to the reseller?
____________________
(\/) Y S T A L

mmmmm, pasta.



Sounds like the silicone spacer under the sink might be interfering. Is it still oriented at a 45-degree angle? AOpen denies this causes trouble, but of course it does.


-j